It’s time to consider how a lack of data collection about some individuals can negatively affect their quality of life

It’s become a popular talking point to list all the risks of data collection, whether it be privacy and surveillance or the lack of transparency that can come with data ownership. But rather than stay bogged down in the potential risks, it’s time to consider how a lack of data collection about some individuals and communities can negatively affect their quality of life.

In today’s digital economy, one significant barrier to opportunity is the data divide, the gaps between the data haves and the data have-nots, and the social and economic inequalities resulting from this lack of data collection and use.

Closing the data divide needs to be a policy priority in the United States to drive robust and equitable growth in the digital economy. Data has become invaluable in today’s economy, where the extent to which individuals and communities can collect data and put it to productive use helps determine everything from health outcomes to public safety and economic growth.

Unfortunately, many have jumped on the bandwagon of criticizing data-driven decisionmaking as too biased. The truth is that some data-driven services don’t work optimally for some people and groups, especially those from historically underrepresented communities because there is oten insufficient data to train these systems. Many of these same critics also argue that data collection is too intrusive, keeping solutions to this problem out of reach. Without prioritising data equity, the United States will continue to perpetuate digital inequalities and miss out on the opportunity for impacful societal change.

Insufficient representation in data poses a serious barrier to many communities and their ability to benefit from data-driven innovation and participate in the data economy. While some individuals are treated with precision medicine and atend schools powered by learning analytics, others make decisions based on incomplete or inaccurate information about themselves, their families and their communities.

These divides manifest in many ways, from demographic and geographic data gaps to inequitable data systems. That means specific characteristics about your background or where you live determine your ability to benefit from data-driven services and whether the necessary data systems and infrastructure exist.

For example, many Americans have unnecessarily low or inaccurate credit scores due to the data infrastructure for financial services.

Credit bureaus oten determine someone’s risk and qualification for loans and other services based on information about financial borrowing and repayment history from traditional financial institutions. But this leaves out key forms of “novel” or alternative data, like on-time rent or utility payments, and even information about cash flow in a bank account.

Similarly, older or underfunded health data infrastructure restricts many patients, providers and researchers in their understanding of individual and community health. Health care lags behind other sectors in updating technologies for the digital era. For example, although 90% of nonfederal acute care hospitals use certified electronic health records (EHR) technology, just 55% use the systems to exchange patient data and 73% have challenges exchanging patient information across different EHR systems.

So, depending on where one lives, if they receive testing or care in one health system, that information doesn’t always transfer between systems, leaving patients with incomplete or inaccurate records. Incomplete EHRS mean less-accurate diagnoses and treatments.

American Indians and Alaska Natives continue to be undercounted in federal statistics. This data gap affects federal funding for digital literacy and broadband access on rural and tribal lands. While the Federal Communications Commission and National Telecommunications and Information Administration support programs to bring broadband access to Native lands, government officials lack the necessary data to understand the scope of the issue and oten allocate resources ineffectively.